08.01.2013 Views

Back Room Front Room 2

Back Room Front Room 2

Back Room Front Room 2

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

16<br />

ENTERPRISE INFORMATION SYSTEMS VI<br />

problems by principals. This can trigger an iterative<br />

reconsideration of the current solution space and its<br />

anomalies resulting in a process called problem<br />

blossoming 15 . This iterative process can change the<br />

contents, and hence, the structure, of the current<br />

solution space (St) as well as the problem space (Pt).<br />

This process may have to be iterated as long as new<br />

affected areas of St are being discovered and the<br />

corresponding anomalies, and resulting problems,<br />

are constructed and organized into the current<br />

problem space. Once complete, or prematurely<br />

stopped by a principal with standing due to the fear<br />

of endless search, the resulting problem set is called<br />

Pt. 16<br />

3.5 Proposed Solution<br />

A Proposed Solution, denoted as St+1, forms a new<br />

subspace of the solution space. A proposed solution<br />

by definition implies the reconciliation of St to Pt. In<br />

other words, each part of a proposed solution must<br />

be reconciled with one or more problems in Pt until<br />

all of the problems in Pt are addressed. The process<br />

of reconciliation, changing St into St+1 17 by solving<br />

for Pt, is called Solution Space Transformation.<br />

Finding this mapping forms the heart of RE. It<br />

involves specifying a mapping from a current<br />

solution space into a future solution space that is<br />

contextualized, or warranted, by the chosen set of<br />

problems. In other words, the analyst's job is at the<br />

intersection of the two solution spaces (along with<br />

technologies embedded in them) and the problem<br />

space. During this reconciliation process, constraints<br />

are seen as limitations of current organizational<br />

resources as well as limitations concerning the future<br />

IS, including people, artifacts, rules, processes and<br />

the like.<br />

It is a custom desire in the RE literature to find an<br />

optimum path from St to St+1. This is, however,<br />

seldom the case in any given requirements analysis<br />

effort, because 1) the prospect of attaining global<br />

solutions is quite remote due to changing and<br />

shifting needs and goals of the principals, problem<br />

blossoming etc, and 2) because system analysts<br />

cannot locally foresee the impact of the chosen<br />

solution spaces or the difficulty of getting there due<br />

to their cognitive and resource limits. The task of the<br />

analyst is, instead, to find a traversable path from a<br />

current solution space to a new one that meets<br />

sufficiently the requirement of removing observed<br />

problems (Haumer, Heymans et al., 1999). This<br />

needs to be accomplished also by identifying<br />

problems that will arise during the process of<br />

transformation.<br />

A necessary outcome of the solution space<br />

transformation is to transform, and possibly expand,<br />

the local solution space, St. Transforming St means<br />

not only changing, and hence a likely expanding<br />

some principals’ technical capability. It also means a<br />

changing, and presumptively by expansion, the<br />

organizational capability within the solution space.<br />

Hence, an expansion of St can reveal previously<br />

unavailable, but now realizable opportunities. The<br />

process can even expand a general solution space,<br />

and thus demonstrate organizational learning and<br />

innovation in the sense that new solution “frames”<br />

have been created (Lyytinen, Rose et al., 1998a) 18 .<br />

3.6 Redefining Requirements<br />

Per our analysis, RE activity involves always a<br />

deliberate construction of an ecology that consists of<br />

two solution spaces and a problem space 19 . The<br />

objective of RE is to reconcile all essential aspects<br />

of the current solution space with regard to a<br />

problem space thus producing a specification for a<br />

particular solution space that can be achieved at<br />

some future time point t+x 20 . It is expected that this<br />

will mitigate or eliminate the identified problem<br />

space (though naturally this cannot be guaranteed).<br />

Due to the discovery of goals, problem blossoming<br />

and dynamics of the solution spaces, this is an<br />

iterative process: new information on both the<br />

solution space and the problem space is continually<br />

discovered, and consequently decisions need to be<br />

continually made to re-state both the solution space<br />

and the problem space in the direction of<br />

reconciliation. The RE specification is thus an<br />

outcome of a co-evolutionary process of discovery<br />

and decision, in which both the solution space and<br />

the problems space are iteratively constructed. This<br />

process is influenced by many constraints arising<br />

from the environment itself (e.g., physical laws,<br />

technical choices, legal considerations, institutional<br />

influences, organizational goals and capabilities,<br />

market forces). But, at the bottom, it remains a social<br />

process of negotiation and inquiry that is constrained<br />

by bounded rationality and limited organizational<br />

resources.<br />

At this point of our treatise we can contrast this<br />

definition of RE with the “received” definition that<br />

is common to the requirements engineering<br />

literature. As previously stated a requirement as per<br />

this literature is:<br />

1. A condition or capability needed by a user to<br />

solve a problem or achieve an objective.<br />

2. A condition or capability that must be met or<br />

possessed by a system or a system component to<br />

satisfy a contract, standard, specification, or<br />

other formally imposed document.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!